CONTACT-02 Highlights: Immunotherapy, Biomarkers, and the Need for Better Treatment Options

A roundtable discussion, moderated by Andrew Laccetti, MD, MS, of Memorial Sloan Kettering Cancer Center, focused on the treatment landscape of metastatic hormone-sensitive prostate cancer (mHSPC) and metastatic castration-resistant prostate cancer (mCRPC), including insights on the integration of recent research and clinical trials, how molecular subtyping affects treatment decisions, the future of care, and more. Dr. Laccetti was joined by Eleni Efstathiou, MD, PhD, of Houston Methodist Cancer Center; Ulka Vaishampayan, MD, of University of Michigan; and Michael Schweizer, MD, of Fred Hutchinson Cancer Center.

In the fifth part of this series, the panel of experts review the CONTACT-02 trial on cabozantinib with atezolizumab for the treatment of mCRPC. The Decipher score utilized in the STAMPEDE trial is also discussed, along with appropriate treatments for patients with metastasis.

View the next segment of this roundtable series: Innovative Approaches in Prostate Cancer: From BiTEs to Estradiol Patches

_

Dr. Laccetti:
I think biomarkers are in great need in our field, and there was some very interesting data presented at the meeting around the Decipher score as a potential informative tool for triplet therapy selection with a retrospective analysis of the STAMPEDE trial, specifically the arm comparing chemotherapy plus docetaxel with ADT monotherapy.

So not a direct resource from triplet therapy data, but the researchers suggested that high Decipher scores prognosticated for improved response to chemotherapy.

Do we think this level of data is strong enough to utilize Decipher as a potential tool to determine who should receive triplet therapy?

Dr. Efstathiou:
A small number of samples were tested; 160 out of several thousands for the Decipher score. I like to be more granular than that; it might not require a score. It might boil down to just a few markers, such as ki-67, plus a couple of others.

But this data is a good start. They had previously published some data with regard to the proliferation index along with a couple of other markers. I think it is upon us now moving forward to actually have a perspective validation of the score itself.

Even if it’s even more simple, something that we can do in our old pathology labs with a

However, something such as immunohistochemistry might be more simple. Let’s not take away from the great value of doing something simple and not costly within our own labs everywhere in the community as long as we are able to have a validated assay.

So I would vote for that, but we need to work together towards that.

Dr. Vaishampayan:
I would say that we are way overdue for Oncotype DX for prostate cancer. This Decipher data may be a start, but I don’t know how much it supersedes Gleason scores, for instance. Typically, my application of Decipher has been for those intermediate-risk patients where you’re totally on the fence and there’s a big spectrum of outcome.

Decipher shows some inkling of being a predictive marker for chemotherapy benefit, and hopefully we can build on that with multiple other markers that have been individually reported, such as ones that tell us the biology of the disease and how aggressive it’s going to be.

I think eventually it’s a panel of biomarkers that will come together that will tell us whether to use chemotherapy or not.

Dr. Schweizer:
Yeah, I agree with what’s been said. The data is intriguing and it’s great to see a molecular biomarker that actually looks like it could be predictive. I’m not going to say it is right now, but the data makes you ask that question. I think we just need prospective validation.

To your point, there are trials being planned now that are looking at not testing triplet therapy versus docetaxel, but triplet therapy versus an ARPI alone. I think the critical question is, can we use something like Decipher to figure out who needs triplet therapy versus who gets by with just something like darolutamide or abiraterone added on to a standard ADT?

Hopefully with time we’ll have answers to those questions. I think the other thing that was interesting to me was that the researchers saw high decipher scores in patients both with high-volume disease – which is how we typically identify patients we think need chemotherapy – but also in low-volume patients too. So it sort of speaks to that being a more precise and better biomarker, but we need more data and studies to prove it.

Dr. Efstathiou:
It’s what I mentioned earlier – the volume is just a minute in time – and it’s such a dynamic disease, you may be picking it up accidentally.

Dr. Schweizer:
Biology is part of the story in terms of mediating volume, but it’s not the whole thing. Decipher actually starts getting at the biology and that’s what we really need.

Dr. Laccetti:
Exactly. Prospective validation trials are what we really need, and those are big, expensive, hard studies to do, but there’s no way around it.

Dr. Vaishampayan:

Also, the bigger point is that we don’t have great surrogates or interim endpoints. If you’re waiting for overall survival, we have to wait a decade or so for follow-up. Getting better surrogates that would give us that interim signal would also be an important thing.

Dr. Laccetti:
I’ll open the question to the panel – are there other studies from the ESMO meeting that you guys think warrant discussion?

Dr. Efstathiou:
Well, you’re setting me up for trouble here, but I think we have to mention CONTACT-02. We got a final analysis. It’s a great study that’s trying to introduce immunotherapy in combination with TKIs, namely atezolizumab plus cabozantinib in men with metastatic CRPC that have aggressive phenotypes, which was selected with clinical criteria.

A lot of us were concerned about the control because having selected for such an aggressive phenotype, you would like to see in the control potentially cytotoxic chemotherapy. But the researchers went from abiraterone to enzalutamide, so these patients had been previously exposed. The data had some positive readouts to it that mainly had to do with men with liver metastasis.

This presentation reaffirmed that if there is value, it is in a subset of patients. Are we ready to get it approved for those men with liver metastasis? We hope the agency and the company are going to discuss it and it will be one more thing that we have for these patients.

Is it enough to make us satisfied that we have something now for these men? No, but we can give them some more hope for another treatment option.

If I were to put my money on something, it would be looking at BiTEs coming through new agents that might be addressing more from the immunotherapy perspective.

Dr. Vaishampayan:
I think one thing that CONTACT-02 highlighted is that this patient population deserves to be addressed differently and that most of these patients are a minority in a lot of our clinical trials.

Focusing on this big unmet need of dismal prognosis in the patient population is what I like about the study and they obviously got a fair amount of patients who had measurable disease. What I don’t like is the lack of prior chemotherapy treatment. So in my management, am I likely to sequence this combination of cabozantinib and atezolizumab prior to chemotherapy?

I find it unlikely unless there is a patient who is specifically not a chemotherapy candidate.

Dr. Schweizer:
Yeah, I agree with what everybody said. I think the issue here is that in the patients who maybe benefit from this combination, they’re the exact patients who really need chemotherapy, such as patients who have liver metastasis. I would not feel comfortable using this.

If you want to redo a study to prospectively show that patients with liver metastasis benefit from this combo, probably in a post-taxane setting, I would welcome that. I think that would be an important study and somebody should do it. Do I think it’s going to happen? Probably not, but I’d like to see more data to actually prove that this combo actually has real benefit there.

We’re seeing a slight progression-free survival advantage across the pre-specified study population and no OS benefit at all.

Dr. Efstathiou: So you’re essentially saying that it should have been a physician’s choice setup where half the patients got docetaxel.

Dr. Schweizer:
Yeah. that would have been more reasonable. I think that you probably still would be asking the question because if I recall, I think the subset with liver tests was like 20% of the overall study. It was a small subset. It’s tough because you’ve got a sort of a post-hoc analysis where you might be seeing a signal there, but that alone is not enough to change practice. You need to have prospective data with predefined endpoints in the population of interest to prove it helps.

Dr. Vaishampayan:
I would argue though that chemotherapy outcomes in this patient population are also very dismal. There’s no basis for chemotherapy in this.

Dr. Schweizer:
Absolutely. And that’s why we need studies, right? And you’re right, we need to have trials that are geared towards these high-risk patients to figure out how to treat it better. It’s an unmet need for sure.

Dr. Vaishampayan:
We need additional options which may be what this combination provides and that we don’t have currently.

Dr. Laccetti:
Absolutely. And the other point I think that needs to be made is that I think toxicity profile needs to be respected. Although the presentation from yesterday highlighted upon it, I think those of us that manage kidney cancer with cabozantinib or nivolumab as monotherapy understand that this is not an easily-tolerated drug for a lot of patients.

And for men who often have symptoms from their cancer, adding on another agent can detract from the quality. We really need to have firm data to support that if that’s warranted.

Dr. Vaishampayan:
The lack of myelosuppression is attractive though, when compared to chemotherapy, because a lot of our patients with really diffuse bone metastasis – even marrow metastasis – have a tough time with blood counts.

Dr. Efstathiou:
I hope and wish that they will also show us some molecular data, because you want to tease out which are the DDRs within these patients.